Currently, there are few codecs that can compress and transmit, for example, a video image of 4K2K size (hi-vision (registered trademark): a quadruple of a resolution of 1920×1080) or a super hi-vision size (16 times of hi-vision) used in digital cinemas, and their costs are high. Accordingly, there is the demand for partitioning a video image into four or 16 images and transmitting the video images by using a plurality of cost-effective hi-vision codecs when a video image of 4K2K size or super hi-vision size (hereinafter referred to as a super hi-vision video image) is transmitted. There is also the demand for implementing a transmission of such a super hi-vision video image by using an Internet Protocol (IP) network that needs less line cost. Moreover, there is the demand for respectively transmitting right and left video images of a 3 Definition (3D) video image with hi-vision as a similar transmission of partitioned images.
When partitioned video images are transmitted by using a plurality of devices, decoders need to synchronize and play back received frames to be play backed of video images of encoders. However, when operations are simply performed such that a video image is partitioned, and partitioned video images and audio are compressed by a plurality of encoders and transmitted in an IP network, and the output video images are synthesized by decoders, the outputs of the partitioned video images of the decoders cannot be synchronized at the same timing. This is because encoding and decoding delays of the encoders and the decoders, and a transmission delay of the network are different. Accordingly, a technique for equalizing delays caused by transmissions between all encoders and decoders by multiplexing and transmitting data corresponding to partitioned video images output from the encoders into one stream, and by demultiplexing the stream on a receiving side is known (for example, see Patent Documents 1 to 3).
As another technique, an encoding/decoding system including, for example, an encoding system configured with a plurality of encoding devices, and a decoding system including a plurality of decoding devices that display image data of one screen by decoding the image data transmitted from the encoding system is known. In such an encoding/decoding system, the encoding system encodes, from the image data of one screen, a plurality of pieces of partitioned image data, the number of which is equal to that of the encoding devices, with the plurality of encoding devices, and transmits the partitioned data to the decoding system. The plurality of encoding devices respectively calculate a time by adding a maximum encoding delay time that can occur at the time of encoding performed by each of the encoding devices to a value of an STC counter indicating a time when each of the encoding devices captures the partitioned image data. The plurality of decoding devices calculate a time by adding stream fluctuations and a maximum possible value of a decoding delay time to the time calculated by the encoding apparatus, and output the image data to a synthesis unit for synthesizing image data at the calculated time. Such a method aims at properly displaying a moving image signal having a high resolution (for example, see Patent Document 4).
Patent Document 1: Japanese Laid-open Patent Publication No. H11-239347
Patent Document 2: Japanese Laid-open Patent Publication No. H10-234043
Patent Document 3: Japanese Laid-open Patent Publication No. H8-79701
Patent Document 4: Japanese Laid-open Patent Publication No. 2008-166862